Back

Odor preference maps to cohesive transcriptional domains in the olfactory bulb

Mayseless, O.; Navajas Acedo, J.; Wan, Y.; Hahaut, V.; Picelli, S.; Friedrich, R. W.; Schier, A. F.

2026-06-16 neuroscience
10.64898/2026.06.11.731560 bioRxiv
Show abstract

Animals classify odors along behaviorally relevant dimensions, yet how the olfactory bulb, the first stage of olfactory processing, is organized and how this organization relates to odor classification remains unclear. Here, we identify cohesive transcriptional domains within the larval zebrafish olfactory bulb whose differential functional recruitment reflects odor preference. Using a flow-based two-choice assay, we characterized behavioral preferences across chemically diverse odors. Mapping odor-evoked neuronal activity patterns revealed that similarly preferred odors converge onto shared spatial activation patterns in the olfactory bulb. Whole-mount multiplexed spatial transcriptomics revealed that projection neuron and interneuron subtypes form cohesive spatial domains. Mapping odor-activated neurons onto these domains identified preference-biased recruitment of specific neuronal subtypes. Targeted ablation of a TH+ interneuron population associated with attraction selectively impaired attraction while sparing aversion. Together, these findings reveal that the olfactory bulb is organized into spatially cohesive, transcriptionally defined domains and that selective recruitment of specific subtypes is linked to odor preference. Key findingsO_LIChemically diverse odorants elicit behaviors reflecting olfactory preference. C_LIO_LITranscriptionally defined neuronal subtypes form cohesive spatial domains in the olfactory bulb. C_LIO_LIOdorants of opposing preference recruit distinct transcriptional domains in the olfactory bulb. C_LIO_LIA TH+ interneuron population is selectively required for odor attraction but not aversion. C_LI

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.